aryehof

> … hard for me to understand Google's redefinition of "world models"

I’ve also wondered what they mean by “world model”, whereas I interpret them as being about predicting outcomes - something an LLM cannot reliably do.

Is this new model an LLM or a predictive outcome model (as you say a la Yann Lecun), or some hybrid?

You can’t control or enable your children …

You can, however, be a good role-model.

> Code review does two jobs: 1. Find and fix issues before they ship 2. Build shared context and codebase understanding across the team

So code review has become newspeak for “testing” and “education”?

Well not really testing. We the users are the testers and it falls to lucky us to really discover and find the errors, edge cases and security issues. The consequence of the rush to continually release…

Is the AI advance in the field of software testing nothing more than “now an AI agent can do a code review?”.

> Computer programming is, fundamentally, about two things: 1. Problem-solving using computers 2. Learning to control complexity while solving these problems

Pretty sure your forgetting more than half of computing…

… 3. The modeling of external systems into code.

Modeling isn't solving a problem, it’s representing an external systems with a degree of fidelity. That’s different from solving a problem through transformation, albeit far harder to teach.

Most software that isn't trivial is a big ball of mud. AI is trained on that and produces the same (or worse) - faster.

Are you following more of an “aider” type approach, or something different?

When I was a child, schools didn’t need to communicate on a daily, weekly (or even monthly) basis with parents. How times change. Heck, we went for a whole school day without having to talk to Mummy or Daddy. So this agent is technology to help the mess enabled by technology?

As a zero-shot classifier, I expect that effectiveness is dependent on the data trained upon.

Jev input … > Unstructured data (e.g. text) with an emphasis on structured program state.

What pre-training data/model is Jev based on? Surely result effectiveness is dependent (outside of one’s own input as “state”) on that?

> Without alignment, further improvements in capability turn LLMs into wanton felony generators

Isn’t “full” alignment and guardrails a task that can never be generally achieved? One will always discover and realize the need for new guidelines and guardrails? What about vague, incomplete, inconsistent nuances and generalizations - that make this a losing battle?

I suggest ultimate safeguards need to be outside the models?

Are the Silo shots too dark to see, like when attempting to watch it via an Apple sub?

Perhaps a better overview is from Reuters …

https://www.reuters.com/business/meta-launches-ai-agent-that...

From the article:

- … rolled out … despite internal concerns that the technology mismanages its access to sensitive personal data

- The product will be available only in the U.S.initially

- A basic version of the agent will be available for free, while Meta will offer subscriptions priced at $20 a month and $100 a month for heavier usage

- People can opt out of their interactions being used to train Meta's AI models.

- Others flagged serious security flaws, like an agent routing around guardrails …

On reading the title, the first thing that came to mind was “a good book”.

Albeit, that I view many of the things in the list as “ideals” or “values”, rather than “luxuries”. The greatest thing one can aspire to in terms of an ideal, is to have “peace of mind” in my opinion.

So true.

> I am kind of amazed how negative the comments are here, especially on HN.

Sadly, I attribute this to a decline in having “good will” towards others, and an increase in cynicism, prejudice, and meanness.

> Build a well architected monolith and be super strict on single purpose and keeping modules separated from each other

Isn’t the question how to keep likely dependent modules separated from each other in a monolith? You say “single purpose”, but how does that admonition work in the reality of a complex problem domain?

> Democracy is "everyone gets a vote"

It is that we are governed by those chosen by us to represent us. In a perfect democracy the government is (the majority of) us. Hence the saying … “in a democracy the people get the government they deserve”.

> We will be building architectural blueprints for software "buildings" same as Civic Engineers and Architects do for bridges and towers.

If only we knew how. All we seem to know formally given popular discourse is languages, frameworks, tools, protocols and algorithms.

Thinking about how to move beyond functional decomposition and structured design to enable those “blueprints” isn’t even on the agenda?

> But what really gets me is the model often tries to make assumptions for me that I didn't specify in the prompt.

This is a problem with your instructions, your specification. An LLM isn't a mind reader. It will attempt to succeed regardless of missing requirements and ambiguity.

I really like the UI. That I prefer to keep my Kindle “books only” is the only reason I might not use inkfeed.

I typically use hckr news (https://hckrnews.com/) set to show only the top 10 or top 20. It also has the advantage of showing those i missed overnight or earlier.

> hit hide on the domains that I never want to see anything from

I thought hide just hid that particular entry. Have I been wrong all this time?

One of the authors is a Google “Evangelist”. A “Chief” one to boot. Why would you expect different?

> I wonder though if we are asking the right question

I also wonder. Is there no further design thought, research or discussion on how to better modularize a program to address complexity - in an LLM world?

Instead, we still can’t move beyond arguing about languages …

For me on a paid plan, the effort indicator was hidden and the model was 5.5 instant. I had to press the + button to select “think harder” before the dial that allowed Sol medium or high to be selected to show.

It made me wonder how many paid subscribers realize they are using the same 5.5 instant model as free users by default. A dark pattern or oversight?

> … with a completely straight face, that they expected 10-100x productivity improvements.

A great alternative future to look forward to - 1 employee expected to do the work of 10-100. Those others fired to save money.

The industry has vast (and increasing) oversupply of “programmers” versus diminishing demand. Add to this, the adoption of AI.

> Overall, I'd like to understand those who have a positive outlook on design and software engineering as a career.

I think until the market better achieves some equilibrium, there is no way general software programming (sorry “engineering”) should be considered as a career. That said, there will always be opportunities in particular markets or specialties.

Sadly interviews are really aptitude and personality tests, perhaps knowledge of leetcode, tools, frameworks and languages.

In an industry that loves calling itself "engineering", it seems nothing further of value can be established?

[delayed]

Interesting. What apps have you built? How/where is state maintained?

If quality isn't valued, then your only alternatives are to try to change that (good luck), accept it, or look for somewhere that does?